Breast cancer is a heterogeneous disease. Despite the establishment of standardized histoclinical criteria, it remains difficult to decide whether to treat patients with node-negative cancer with or without adjuvant chemotherapy, and to recognize among the patients who receive adjuvant chemotherapy those who will benefit and those who will not benefit from standard anthracyclin-based protocols. Prognostic classification can benefit from large-scale expression profiling. Using immunohistochemistry on tissue microarrays, Jacquemier et al. monitored the expression of 26 selected proteins in more than 1,600 cancer samples from 552 consecutive patients with early breast cancer. Using a supervised method, they identified a set of 21 proteins (top image) whose combined expression significantly correlated to metastasis-free survival (MFS) in a learning set (368 patients) and in an independent validation set (184 patients). Among the 552 patients, the 5-year MFS was 90% for patients classified in the "good-prognosis class" and 61% for those classified in the "poor-prognosis class" (p<0.0001) (bottom image). In multivariate analysis, the 21-protein set was the strongest independent predictor of clinical outcome. These results show promise to assess breast cancer heterogeneity and prognosis in stage I, II, or III disease. For details, see the article by Jacquemier et al. on page 767 of this issue.